basic tests for picklist functionality

src/sourmash/sig/picklist.py

@@ -10,8 +10,8 @@
 preprocess['md5'] = lambda x: x
 
 # identifier matches/prefix foo - space delimited identifiers
-preprocess['ident'] = lambda x: x.split(' ')[0].split('.')[0]
-preprocess['ident.'] = lambda x: x.split(' ')[0]
+preprocess['ident.'] = lambda x: x.split(' ')[0].split('.')[0]
+preprocess['ident'] = lambda x: x.split(' ')[0]
 
 # match 8 characters
 preprocess['md5prefix8'] = lambda x: x[:8]

tests/test_cmd_signature.py

@@ -1110,6 +1110,161 @@ def test_sig_extract_7_no_ksize(c):
     assert len(siglist) == 3
 
 
+def test_sig_extract_8_picklist_md5(runtmp):
+    # extract 47 from 47, using a picklist w/full md5
+    sig47 = utils.get_test_data('47.fa.sig')
+    sig63 = utils.get_test_data('63.fa.sig')
+
+    # select on any of these attributes
+    row = dict(exactName='NC_009665.1 Shewanella baltica OS185, complete genome',
+               md5full='09a08691ce52952152f0e866a59f6261',
+               md5short='09a08691ce5295215',
+               fullIdent='NC_009665.1',
+               nodotIdent='NC_009665')
+
+    # make picklist
+    picklist_csv = runtmp.output('pick.csv')
+    with open(picklist_csv, 'w', newline='') as csvfp:
+        w = csv.DictWriter(csvfp, fieldnames=row.keys())
+        w.writeheader()
+        w.writerow(row)
+
+    picklist_arg = f"{picklist_csv}:md5full:md5"
+    runtmp.sourmash('sig', 'extract', sig47, sig63, '--picklist', picklist_arg)
+
+    # stdout should be new signature
+    out = runtmp.last_result.out
+
+    test_extract_sig = sourmash.load_one_signature(sig47)
+    actual_extract_sig = sourmash.load_one_signature(out)
+
+    assert actual_extract_sig == test_extract_sig
+
+
+def test_sig_extract_8_picklist_name(runtmp):
+    # extract 47 from 47, using a picklist w/full md5
+    sig47 = utils.get_test_data('47.fa.sig')
+    sig63 = utils.get_test_data('63.fa.sig')
+
+    # select on any of these attributes
+    row = dict(exactName='NC_009665.1 Shewanella baltica OS185, complete genome',
+               md5full='09a08691ce52952152f0e866a59f6261',
+               md5short='09a08691ce5295215',
+               fullIdent='NC_009665.1',
+               nodotIdent='NC_009665')
+
+    # make picklist
+    picklist_csv = runtmp.output('pick.csv')
+    with open(picklist_csv, 'w', newline='') as csvfp:
+        w = csv.DictWriter(csvfp, fieldnames=row.keys())
+        w.writeheader()
+        w.writerow(row)
+
+    picklist_arg = f"{picklist_csv}:exactName:name"
+    runtmp.sourmash('sig', 'extract', sig47, sig63, '--picklist', picklist_arg)
+
+    # stdout should be new signature
+    out = runtmp.last_result.out
+
+    test_extract_sig = sourmash.load_one_signature(sig47)
+    actual_extract_sig = sourmash.load_one_signature(out)
+
+    assert actual_extract_sig == test_extract_sig
+
+
+def test_sig_extract_8_picklist_ident(runtmp):
+    # extract 47 from 47, using a picklist w/full md5
+    sig47 = utils.get_test_data('47.fa.sig')
+    sig63 = utils.get_test_data('63.fa.sig')
+
+    # select on any of these attributes
+    row = dict(exactName='NC_009665.1 Shewanella baltica OS185, complete genome',
+               md5full='09a08691ce52952152f0e866a59f6261',
+               md5short='09a08691ce5295215',
+               fullIdent='NC_009665.1',
+               nodotIdent='NC_009665')
+
+    # make picklist
+    picklist_csv = runtmp.output('pick.csv')
+    with open(picklist_csv, 'w', newline='') as csvfp:
+        w = csv.DictWriter(csvfp, fieldnames=row.keys())
+        w.writeheader()
+        w.writerow(row)
+
+    picklist_arg = f"{picklist_csv}:fullIdent:ident"
+    runtmp.sourmash('sig', 'extract', sig47, sig63, '--picklist', picklist_arg)
+
+    # stdout should be new signature
+    out = runtmp.last_result.out
+
+    test_extract_sig = sourmash.load_one_signature(sig47)
+    actual_extract_sig = sourmash.load_one_signature(out)
+
+    assert actual_extract_sig == test_extract_sig
+
+
+def test_sig_extract_8_picklist_ident_dot(runtmp):
+    # extract 47 from 47, using a picklist w/full md5
+    sig47 = utils.get_test_data('47.fa.sig')
+    sig63 = utils.get_test_data('63.fa.sig')
+
+    # select on any of these attributes
+    row = dict(exactName='NC_009665.1 Shewanella baltica OS185, complete genome',
+               md5full='09a08691ce52952152f0e866a59f6261',
+               md5short='09a08691ce5295215',
+               fullIdent='NC_009665.1',
+               nodotIdent='NC_009665')
+
+    # make picklist
+    picklist_csv = runtmp.output('pick.csv')
+    with open(picklist_csv, 'w', newline='') as csvfp:
+        w = csv.DictWriter(csvfp, fieldnames=row.keys())
+        w.writeheader()
+        w.writerow(row)
+
+    picklist_arg = f"{picklist_csv}:nodotIdent:ident."
+    runtmp.sourmash('sig', 'extract', sig47, sig63, '--picklist', picklist_arg)
+
+    # stdout should be new signature
+    out = runtmp.last_result.out
+
+    test_extract_sig = sourmash.load_one_signature(sig47)
+    actual_extract_sig = sourmash.load_one_signature(out)
+
+    assert actual_extract_sig == test_extract_sig
+
+
+def test_sig_extract_8_picklist_md5_short(runtmp):
+    # extract 47 from 47, using a picklist w/full md5
+    sig47 = utils.get_test_data('47.fa.sig')
+    sig63 = utils.get_test_data('63.fa.sig')
+
+    # select on any of these attributes
+    row = dict(exactName='NC_009665.1 Shewanella baltica OS185, complete genome',
+               md5full='09a08691ce52952152f0e866a59f6261',
+               md5short='09a08691ce5295215',
+               fullIdent='NC_009665.1',
+               nodotIdent='NC_009665')
+
+    # make picklist
+    picklist_csv = runtmp.output('pick.csv')
+    with open(picklist_csv, 'w', newline='') as csvfp:
+        w = csv.DictWriter(csvfp, fieldnames=row.keys())
+        w.writeheader()
+        w.writerow(row)
+
+    picklist_arg = f"{picklist_csv}:md5short:md5prefix8"
+    runtmp.sourmash('sig', 'extract', sig47, sig63, '--picklist', picklist_arg)
+
+    # stdout should be new signature
+    out = runtmp.last_result.out
+
+    test_extract_sig = sourmash.load_one_signature(sig47)
+    actual_extract_sig = sourmash.load_one_signature(out)
+
+    assert actual_extract_sig == test_extract_sig
+
+
 @utils.in_tempdir
 def test_sig_flatten_1(c):
     # extract matches to several names from among several signatures & flatten